The aim of this proposal is to develop techniques of NMR blood flow imaging of tumors which will have value in clinical oncology. We will investigate spatially resolved flow quantification by means of nuclear magnetic resonance (NMR) imaging. The research effort will be focused in the areas of: liquid flow through porous media, image reconstruction algorithms, methods, and systems necessary for the spatially resolved velocity variance imaging and vascular volume determination from MRI images. A special case of velocity variance occurs in the case of random directional flow. The type of incoherent flow which will be investigated in this proposal differs from molecular diffusion in that here incoherence arises either from the tortuousity of the flow channels or the existence of a velocity spectrum within a voxel. In order to test our hypothesis we will initially use porous materials to simulate tissues with tortuously distributed vessels. The method, however is general enough to handle any degree of tortuosity. At a later time during the investigation isolated animal organs will be used to test the techniques developed on phantoms. Ultimately, we will image transplantable tumors in the rabbit hind lamb. We will initially mimic therapeutically induced flow changes with starch microspheres. Later, we will obtain serial images following therapy with chemotherapeutic agents active against these tumors. We will use perfluoroctylbromide (PFOB) containing F-19 as a contrast material to eliminate the stationary background tissues. The localization of PFOB in the vessels will be achieved by fluorine (F-19) imaging. Since PFOB NMR spectrum exhibits large chemical shifts we will develop techniques to deal with the artifacts arising from such shifts. The successful completion of this proposal would enable scientists and clinicians to characterize perfusive blood flow accurately. Such information would aid in the understanding of blood flow through tumors. Accurate measurement and imaging of blood flow through tumors may be useful in the non-invasive monitoring of tumor response to anti-cancer therapy.